The vpu protein of human immunodeficiency virus type 1 plays a protective role against virus-induced apoptosis in primary CD4(+) T lymphocytes

Abstract

Previous data revealed that primary cultures of peripheral blood mononuclear cells (PBMCs) were killed by apoptosis at higher rates after infection with two CRF01_AE primary isolates of human immunodeficiency virus type 1 (HIV-1) than after infection with five other CRF01_AE primary isolates, five subtype B primary isolates, and two subtype B laboratory strains. Here, we show evidence that mutations at the vpu gene which were exclusively identified only in the two CRF01_AE isolates mentioned above are involved in their abilities to induce massive apoptosis in primary CD4(+) T lymphocytes. The rates of virus production by these two isolates in the culture media of infected PBMCs were lower (the same as those of the other CRF01_AE isolates) than those of the subtype B isolates. To confirm the correlation between the higher apoptosis-inducing abilities and the mutations at the vpu gene, infectious molecular clone pNL4-3-based vpu mutants were constructed and examined for their apoptosis induction levels. The apoptosis induction levels after introduction of the vpu mutations were greatly increased in primary CD4(+) T lymphocytes. In contrast, the apoptosis induction abilities of these vpu mutants were lower in human T-cell line MT-4. Thus, the Vpu protein of HIV-1 could play a protective role against virus-induced apoptosis in primary CD4(+) T lymphocytes.

FIG. 1.

vpu sequences of five subtype B and seven CRF01_AE primary isolates in addition to the NL4-3 and Ba-L subtype B laboratory strains. The Vpu amino acid sequences of the HIV-1 primary isolates were determined as described in Materials and Methods. The percentages of apoptosis induction in PBMCs on day 7 after infection with these primary isolates are shown at the right (means and standard deviations of triplicate experiments). An asterisk indicates the appearance of a stop codon.

FIG. 2.

Mutations at vpu gene in 95TNIH022 and 95TNIH047. (A) Schematic representations of vpu sequences of 95TNIH022 (amino acid residues 1 to 22, with one amino acid insertion in the TM domain and no cytoplasmic domain) and 95TNIH047 (amino acid residues 1 to 27, with one amino acid insertion in the TM domain and no cytoplasmic domain). As a control, NL4-3 containing wild-type vpu is shown. LTR, long terminal repeat. (B) Western blotting of Vpu protein in PBMCs infected with 95TNIH022, 95TNIH047, and NL4-3. (Left blot) An anti-Vpu antibody recognized the cytoplasmic domain. (Right blot) A similarly prepared blot was reacted with serum from an HIV-1-seropositive individual.

FIG. 3.

Comparison of the kinetics of HIV-1 protein synthesis and the release of virions from PBMCs infected with 95TNIH022 or 95TNIH047 and with NL4-3 as a control. PBMCs were infected with 95TNIH022, 95TNIH047, or NL4-3. After adsorption for 2 h, the infected PBMCs were cultured for 12 days. The HIV-1 particles released into the culture media obtained every 2 days were quantified by the Gag p24 antigen capture ELISA (upper panels). The data shown are the means and standard deviations of triplicate experiments. The viral proteins in the infected cells were visualized by Western blotting (middle panels) and IFA (lower panels) with serum from an HIV-1-seropositive individual. The infected cells used for the experiments shown in the middle and lower panels were derived from one of the triplicate experiments shown in the upper panels. IFA data represent the sequential percentages of infected cells.

FIG. 4.

Construction of NL4-3-based vpu mutants. (A) Three vpu mutants were constructed by introducing three different vpu mutant genes into pNL4-3: the NL4-3-derived gene for the TM region spanning amino acid residues 1 to 21, designated NL4-3(NL-vpu-TM); the 95TNIH022-derived vpu gene, designated NL4-3(022-vpu-TM); and a complete deletion of the vpu gene, designated NL4-3(Δvpu). As a control, wild-type NL4-3 [NL4-3(WT)] was used. LTR, long terminal repeat. (B) Expression of Vpu protein in PBMCs infected with these three vpu mutants and NL4-3(WT), as detected by Western blotting with an anti-Vpu antibody recognizing the cytoplasmic domain and with serum from an HIV-1-seropositive individual as a control antibody.

FIG. 5.

Augmentation of apoptosis induction by NL4-3 in human primary CD4⁺-T-cell cultures after the introduction of vpu mutations. (A) PBMCs (left panel) or a PBMC-derived CD4⁺-T-cell subpopulation (right panel) was mock infected (□) or infected with NL4-3 (▵), 95TNIH022 (○), NL4-3(NL-vpu-TM) (▾), NL4-3(022-vpu-TM) (⧫), or NL4-3(Δvpu) (▪). The apoptosis induction levels were measured every 2 days by TUNEL staining. (B) Simultaneously, the Gag p24 production levels in the culture media were also measured by the antigen capture ELISA. The data shown in panels A and B are the means and standard deviations of triplicate experiments. (C) Sequential percentages of infected cells, as determined by IFA. The data shown are representative of three independent cell cultures.

FIG. 6.

Reduction of apoptosis induction ability of NL4-3 in human CD4⁺-T-cell line MT-4 by the introduction of vpu mutations. MT-4 cells were mock infected (□) or infected with NL4-3(WT) (▵), NL4-3(NL-vpu-TM) (▾), NL4-3(022-vpu-TM) (⧫), or NL4-3(Δvpu) (▪). The apoptosis induction levels were measured every 2 days after infection by TUNEL staining (left panel). Simultaneously, the Gag p24 production levels in the culture media (middle panel) and the sequential percentages of infected cells (right panel) were measured by the antigen capture ELISA and IFA, respectively. The data shown are the means (and standard deviations, for the left and middle panels) of triplicate experiments.

FIG. 7.

Reduction of apoptosis induction ability of NL4-3 in human kidney cell line 293T by the introduction of vpu mutations. (A) 293T cells were mock transfected or transfected with pNL4-3(WT), pNL4-3(NL-vpu-TM), or pUC18 as a negative control in a single-round infection assay. The apoptosis induction levels were measured 3 days after transfection by TUNEL staining, followed by flow cytometric analysis. The data shown are the means and standard deviations of triplicate experiments. (B) (Upper panels) Simultaneously, the Gag p24 production levels in the culture media from the cells transfected with pNL4-3 (▵) and pNL4-3(NL-vpu-TM) (▾) were measured by the antigen capture ELISA. (Middle and lower panels) The viral proteins in the same transfected cells were visualized by Western blotting and IFA, respectively, with serum from an HIV-1-seropositive individual. The data shown in the upper and lower panels are representative of three independent experiments (means and standard deviations).